INDEX

Explanations

terms related to programming constructs or entities within the code

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 transfieras

-0.69

uxxxx

-0.51

 minY

-0.46

+#+

-0.44

styleType

-0.44

ValueGeneration

-0.43

Jereo

-0.42

 getMin

-0.42

>'+

-0.41

 ویکی‌پدی

-0.40

POSITIVE LOGITS

ArgsConstructor

0.56

DispatchToProps

0.52

 tarko

0.46

новништво

0.46

appartamento

0.45

 kaupung

0.44

 cár

0.43

 boste

0.42

 ympä

0.42

SuspendLayout

0.41

Activations Density 0.005%

terms related to programming constructs or entities within the code

No Comments

No Known Activations